臺灣博碩士論文加值系統

近年來，在電子資訊的應用領域□，即時系統的重要性已逐漸增加，其中硬即時的要求更通常是國防、航太、醫學及工廠自動化等即時系統設計上的基本規格。有關硬即時系統的設計，除硬體性能的配合外，在軟體設計上，通常包含工作分割、排程演算法設計及即時作業系統的支援等考量，其中即時作業系統的支援為整體硬即時系統的設計與運作基礎。本篇論文係利用一個精簡且為開放程式碼的即時核心，藉由相關工作排程機制的修改，強化其支援硬即時系統的功能，同時並增加其可調變構型的功能，以提供設計者更大的彈性，最後結合修改後的作業系統核心與一個實際的硬即時系統，藉由完整的系統開發環境以測試整體系統功能，同時並討論系統設計人員設計硬即時系統應考量的因素。

Recently, real-time systems have become more and more important in the field of electronic and information applications. And the requirement of hard real-time has often become the basic specification for the real-time systems used in national defense, space technology, medication application and industrial automation.
Many applications deal with sets of activities or tasks and make them real-time systems. They need an appropriate real-time operating system to support their development and operation.
The basic design criteria for hard real-time systems includes hardware selection, tasks arrangement, scheduling design and support of real-time system. Among them, the real-time operating system is the most critical and essential issue.
uC/OS-II is a real-time operating system with open source code. Because of the characteristics of priority-driven and preemptive for uC/OS-II, it is only used to develop the soft real-time system. So we tried to revise the original kernel and increase its hard real time performance.
In this paper, we discuss how to make uC/OS-II a configurable hard real-time operating system. We verified our proposed solution on a simplified standalone simulation system. And an existing hard real-time system is implemented base on our revised uC/OS-II. The whole system is tested in the integrated development environment. Finally, the design criteria to build a hard real-time system are also evaluated.

1. Introduction.....1
2. Background.....5
2.1 The Hard Real-Time Issues.....5
2.2 The Architecture of the uC/OS-II.....7
2.3 The Operation of the uC/OS-II Kernel.....9
2.4 Shortcomings of uC/OS-II to Support the Hard Real-Time System.....12
3. Proposed Solution.....18
3.1 Revisions of uC/OS-II to Support the Hard Real-Time System.....18
3.1.1 Hard Real-Time Guarantee.....18
3.1.2 Deadlock Check.....22
3.1.3 Real-Time Metrics.....26
3.1.4 Reconfiguration.....27
3.2 Analysis.....30
3.2.1 A Typical Flight Control System.....30
3.2.2 Evaluation.....38
3.3 Design Criteria.....42
3.4 Integrated Development and Verification Requirements.....43
3.5 Experiment Results.....45
4. Conclusions and Future Works.....50
References.....52

[1] Burns, A. and Wellings, A.J., “Real-Time Systems and their Programming Languages,” Addison-Wesley, 1990[2] N. C. Audsley A. Burns M. F. Richardson A. J. Wellings, “Incorporating Unbounded Algorithms Into Predictable Real-Time Systems,” 1993[3] http://www.ucos-ii.com/, Web site of uC/OS-II[4] Victor Yodaiken, “The RTLinux Manifesto”[5] Jean J. Labrosse, “MicroC/OS-II The Real-Time Kernel,” Prentice Hall Inc, 1996[6] Martin Timmerman and Jean-Christophe Monfret, “Windows NT as Real-Time OS?,” Real-Time Magazine, Feb. 1997[7] http://www.ucos-ii.com/prod.htm, Web site of uC/OS-II[8] Iain John Bate, “Scheduling and Timing Analysis for Safety Critical Real-Time Systems,” Department of Computer Science, University of York, 1998[9] Victor Yodaiken and Michael Barabanov, “RTLinux Version 2,” 1999[10] Yu-Chung Wang and Kwei-Jay Lin, “Enhancing the real-time capability of the Linux kernel,” 1998[11] Robert Hill, Balaji Srinivasan, Shyam Pather and Douglas, “Temporal Resolution and Real-Time Extensions to Linux,” 1998[12] http://www.lynx.com/, Web site of LynxOS[13] http://www.qnx.com/, Web site of QNX[14] Pablo J. Rogina and Gabriel Wainer, “New Real-Time Extensions to the MINIX operating system,” August, 1999[15] Wu-Fong Chen, “A Real-Time Flight Control Embedded System Based on uC/OS-II Kernel,” 2000